譯文 |“數(shù)據(jù)貨幣化”戰(zhàn)略的五大要素

來源：數(shù)據(jù)觀-自媒體 時間：2017-02-15 11:31:47 作者：燈塔大數(shù)據(jù)

?當(dāng)今，在認(rèn)知計(jì)算時代下的數(shù)字化商業(yè)模型中，數(shù)據(jù)帶來了新的收入流。如果一個公司能夠高效地利用數(shù)據(jù)，那么認(rèn)知計(jì)算學(xué)就能為其帶來額外的收入流。

?在大數(shù)據(jù)中，我們稱之為“數(shù)據(jù)貨幣化”。數(shù)據(jù)貨幣化已經(jīng)在全行業(yè)中掀起了改革的浪潮，提高了用戶體驗(yàn)，使更精準(zhǔn)的個性化市場和銷售策略成為可能，還有效地防止了詐騙的發(fā)生。

?大數(shù)據(jù)的興起推動了各行各業(yè)的改革，大數(shù)據(jù)在成本優(yōu)化和用戶體驗(yàn)提高方面已經(jīng)顯出了巨大的作用，越來越多的公司發(fā)現(xiàn)大數(shù)據(jù)能夠?yàn)樗麄儙硇碌氖杖肓鳌?/p>

?從銀行業(yè)到電信業(yè)，從能源業(yè)到零售業(yè)，只要手握數(shù)據(jù)，這些公司就能創(chuàng)造出新的盈利點(diǎn)。這些行業(yè)都正在經(jīng)歷著數(shù)據(jù)價值“貨幣化”的過程，通過優(yōu)化數(shù)據(jù)收集和儲存過程獲得了更大的盈利空間。

?麥肯錫全球研究所的《大數(shù)據(jù)研究報告》顯示，在創(chuàng)新、競爭和生產(chǎn)效率的發(fā)展前線上，大數(shù)據(jù)可以為客戶端用戶和企業(yè)端用戶創(chuàng)造7000億美元的價值。想要獲得這一價值，就必須要在技術(shù)、基礎(chǔ)設(shè)施、人力方面有足夠的投入，政府也需要給予一定的支持。

?1、發(fā)現(xiàn)目標(biāo)客戶需求、要求和期望

?在利用數(shù)據(jù)挖掘利潤之前，你必須先找準(zhǔn)目標(biāo)客戶，并列出行業(yè)競爭對手，分析他們成功的原因。

?以樂購（Tesco）公司為例，他們需要關(guān)注零售商和購物商場的運(yùn)營情況，獲取人們的購物活動信息，從而進(jìn)行物流及庫存管理和客戶來源地區(qū)分析，因?yàn)檫@些分析需要基于真實(shí)的客戶行為數(shù)據(jù)。

?2、發(fā)現(xiàn)數(shù)據(jù)集特點(diǎn)——原始數(shù)據(jù)or修正數(shù)據(jù)？內(nèi)部數(shù)據(jù)or外部數(shù)據(jù)？

?數(shù)據(jù)貨幣化并非僅僅是儲存和出售數(shù)據(jù)。數(shù)據(jù)貨幣化對數(shù)據(jù)分析過程、分析結(jié)果和合作伙伴都有一定要求。我們不妨組建一支集中化管理的數(shù)據(jù)科學(xué)隊(duì)伍，與公司企業(yè)合作，分析不同數(shù)據(jù)集特征，探索應(yīng)用案例，引進(jìn)新的業(yè)務(wù)團(tuán)隊(duì)。

?目前IBM與很多零售公司都建立了合作關(guān)系，這些零售公司用Hadoop和Spark整理數(shù)據(jù)，形成供給鏈實(shí)時報告，然后將報告賣給批發(fā)商。值得注意的是，這些數(shù)據(jù)不僅包括銷售點(diǎn)的購買數(shù)據(jù)，還有從銀行處獲得的交易數(shù)據(jù)。

?在Apache Spark和Kafka的幫助下，形成報告只需要數(shù)小時的時間，正確使用可擴(kuò)模型可以將整體收入提高25%。分析這些報告可以為公司在客戶區(qū)分和交叉銷售分析方面提供很大幫助。

?3、技術(shù)的合規(guī)性和合法性問題

?分享數(shù)據(jù)時，人們通常會遇到數(shù)據(jù)被盜用的問題。因此我們應(yīng)該建立明確的問責(zé)機(jī)制和準(zhǔn)入門檻，遵守國家關(guān)于數(shù)據(jù)安全、隱私和自留責(zé)任等方面的政策，以確?？蛻舨粫ξ覀儐适湃危膊粫|發(fā)任何法律法規(guī)的禁區(qū)。公司隱私政策必須言簡意賅、通俗易懂。

?4、數(shù)據(jù)服務(wù)與商業(yè)模型

?要落實(shí)數(shù)據(jù)貨幣化戰(zhàn)略就必須選擇合適的商業(yè)模型，建立有力的戰(zhàn)略聯(lián)盟，找到靠譜的合作伙伴。

?很多公司專門做高級大數(shù)據(jù)服務(wù)。如果這些數(shù)據(jù)公司能夠?yàn)榭蛻籼峁┐罅坑袃r值的數(shù)據(jù)，那么就可以達(dá)到雙贏的結(jié)果。

?5、確立技術(shù)戰(zhàn)略——Hadoop、Spark和IBM Watson數(shù)據(jù)平臺

開源技術(shù)為公司企業(yè)在數(shù)據(jù)貨幣化的發(fā)展中提供了有力支持。越新的數(shù)據(jù)，價值越高。Apache Spark和Kafka等技術(shù)都能為企業(yè)提供快速的實(shí)時數(shù)據(jù)分析，這 種數(shù)據(jù)處理方式和管理方式是前所未有的。

?簡單來說，這些改變都是為了一個結(jié)果——提高數(shù)據(jù)的靈活性。

?理想的大數(shù)據(jù)環(huán)境是由開放標(biāo)準(zhǔn)驅(qū)動的，并且是鼓勵合作的。Hadoop、Spark和IBM Watson等大數(shù)據(jù)平臺可以為數(shù)據(jù)貨幣化戰(zhàn)略奠定堅(jiān)實(shí)的基礎(chǔ)，幫助企業(yè)迅速地實(shí)現(xiàn)數(shù)據(jù)的貨幣化。

英文原文：

?5 key attributes of effective data monetization strategy

?In cognitive computing era, new revenue generation stream has emerged with data at center of the modern digital business model. One of the key capabilities cognitive computing enables for an organization is the ability to generate additional revenue streams by using data effectively.

?In the big data world we call it data monetization. The internal data monetization has already done amazing job at transforming business in all verticals by improving customer experience, enabling more personalized marketing and sales, deterring fraud and so on.

?The emergence of big data has shown to transform professions and industries. We are seeing big data doing wonders with cost optimization and enhancing customer experience. We are increasingly seeing a growing trend among our customers to create new revenue streams with big data.

?Customers ranging from banks, telecommunication providers, energy and utilities companies and retailers have potential to earn new revenues from the vast amount of data they hold. Each of these businesses are experimenting with different ways to monetize the value of the data they gather during their normal operations. Each are expecting to make considerable revenues based upon the difference between the cost of collecting and storing the data, and what the insights and outcomes can be sold for.

?As per the McKinsey Global Institute report on “Big data: The next frontier for innovation, competition, and productivity,” big data can create as much as $700 billion in value to consumer and business end users. Capturing this value will require the right enablers, including sufficient investment in technology, infrastructure and personnel as well as appropriate government action.

?1. Identifying your target customers' needs, requirements and aspirations

?Before you embark on journey to make money out of your data. It is important you profile your target customers, verticals and their parameters for success.

?Case in point is telcos targeting retailers and mall operators with insights about anonymous movement of people throughout the property and surrounding. Delivering store or business catchment analysis based on real behavior, not just proximity to your location.

?2. Identifying data assets—raw and refined, internal and external

?Data monetization is much more than just storing and selling the data. Data monetization is about making revenue out of data enablers like insights, outcomes and partnerships. Companies can benefit from a centralized Data Science team that partners with the business and potential customers by identifying data that differentiates, exploring use cases to solve, and helping to jumpstart business teams.

?One of the customer engaged with us is a retail company who is selling real time supply chain report to merchant wholesalers. The company is using the data from their Hadoop and Spark cluster to generate revenue-driving reports for wholesalers. The key parameter here is blending of purchase data from POS with transaction data from banks. With Apache Spark and Kafka, they run these reports in just hours, and with the scalability models in place they expect to grow this business to 25% of overall revenue. The analytics from these reports help merchants with customer segmentation, cross-sell analytics, and more.

?3. Addressing regulatory and legal issues with technology.

?How you share your data is about balancing needs to innovate against the risk of using your data. Strike that balance with clear responsibilities and pragmatic access, enforce compliance to data security, privacy and retention policies and processes to ensure continued trust by consumers and meet regulatory and legal requirements. Company privacy policies must be clear and well-understood by overall business and technical team. Access should be determined by the use case requirements and priorities.

?4. Data as a service and business model

?Operationalizing your data monetization strategy calls for having the right business model, the right strategic alliances and the right partner.

?The companies are working on driving sophisticated big data as a service business models based on both volumes and values. The win-win business model will be highly influenced by the number of insights business can provide to customers and value those insights can generate for their customers.

?5. Defining the technology strategy—Hadoop, Spark and IBM Watson Data Platform

?The emergence of open source technologies gives tremendous power to organization in this new emerging data monetization space to break even more swiftly. Data provides maximum value when it is fresh. Technologies like Apache Spark and Kafka give real time analysis capabilities to business at lightning speed. This technology has a wholly different approach to data and data management than what we had before. It is the key enabler to the far reaching transformation that is really “big data.”

?In short, these changes all lead back to the simplest of facts in the underlying technology—the agility of data.

?A big data environment that supports collaboration powered by open standards is ideal. IBM Watson Data Platform provides the power of machine learning and cognitive computing based on open source “Apache Spark” to enterprises. Data platforms such as this will form solid foundation for a data monetization strategy and will enable organizations to quickly and easily monetize data.

責(zé)任編輯：陳近梅